spd.interpolate: Interpolation between two symmetric, positive-definite...

In areshenk/spdm: Geometry-respecting analysis of covariance matrices

Interpolation between two symmetric, positive-definite matrices

Description

Function smoothly interpolates between two SPD matrices x and y. The interpolation is parametrized by t, where t=0 returns x, t=1 returns y, and t=.5 returns the mean of x and y

Usage

spd.interpolate(x, y, t, method = "euclidean", ...)

Arguments

x, y	Symmetric, positive-definite matrices
t	Interpolation parameter in [0,infinity)
method	Type of interpolation (see details)

Details

Allowable distance measures are

• "euclidean": Euclidean interpolation (1-t)x + ty

• "logeuclidean": Euclidean interpolation in the tangent space.

• "riemannian": Interpolation along the geodesic path from x to y

areshenk/spdm documentation built on Oct. 2, 2024, 10:39 a.m.